Batch, Stochastic and Mini-Batch Gradient Descent
About this video
Batch, stochastic and mini-batch: three ways to traverse the data.
Batch gradient descent uses all the data at each step, the stochastic version uses a single sample and mini-batch picks an intermediate group, usually 32 to 256 examples. This guide compares the three variants, their cost, their noise and why mini-batch has become the standard for training modern networks.
Video transcript
- 0:01 Full-batch descent uses all the data on every step:
- 0:04 a smooth path, but each step is expensive.
- 0:08 SGD uses a single sample and moves with a lot of noise.
- 0:12 Mini-batch sits in between.
More videos
Explainer
The Computational Graph and Automatic Differentiation
The computational graph and automatic differentiation, in action.
Explainer
Backpropagation: The Step-by-Step Derivation
The backpropagation derivation, factor by factor with the chain rule.
Explainer
Backpropagation: The Intuition
The intuition of backpropagation: the error flows backward through the network.
